Creati ve i ndustri es

Heterogenei ty and connect i on wi t h r egi onal
f i r m entr y
J.P.J. de Jong (EIM Business and Policy Research)
P. Fris (EIM Business and Policy Research)
E. Stam (University of Utrecht)
Zoetermeer, December 14, 2007

This report is published under the SCALES-initiative (SCientific AnaLysis of Entrepreneur-
ship and SMEs), as part of the 'SMEs and Entrepreneurship programme' financed by the
Netherlands Ministry of Economic Affairs.
Most recent EIM reports and much more on SMEs and Entrepreneurship can be found at:

The responsibility for the contents of this report lies with EIM bv. Quoting numbers or text
in papers, essays and books is permitted only when the source is clearly mentioned. No part
of this publication may be copied and/or published in any form or by any means, or stored
in a retrieval system, without the prior written permission of EIM bv. EIM bv does not
accept responsibility for printing errors and/or other imperfections.


Summary 5
1 Introduction 7
2 The nature of creative industries 9
3 Heterogeneity of business processes 15
4 Contribution to regional firm entry 21
5 Discussion 27

Annex I References 29
Annex II Classification of creative industries 31

Creative industries are nowadays central in many policies to stimulate the
economic development of cities, regions and advanced capitalist economies
in general. This paper contributes to the creative industries literature in
two respects. First, we empirically explore if high shares of creative indus-
tries in regions go together with one particular aspect of regional economic
development, namely firm entry rates. Drawing on Dutch trade register
data over a six-year period, it is concluded that at the level of municipali-
ties there is indeed a connection between the share of creative industries
and firm entry, even after controlling for the sizes of municipalities, and no
matter if creative industries are defined broadly or narrowly. Second, the
paper analyses if firms in creative industries are heterogeneous in terms of
business processes and their contribution to regional firm entry. Drawing
on previous work four creative domains are identified: arts, media and en-
tertainment, creative business services and, at the periphery, knowledge
intensive business services. After analysing survey data of 4,746 Dutch
SMEs, we find that firms across these domains are distinct in their use of
the surveyed business practices: innovation, strategy and marketing, and
human resources practices. Especially knowledge intensive services firms
are deviant. For the connection with firm entry rates, it appears that high
shares of firms in the arts and knowledge intensive business services are
significantly connected with regional firm entry rates, while media and en-
tertainment and creative business services remain insignificant. Implica-
tions for practitioners and future research are discussed.

Keywords: creative industries, business processes, innovation, heterogene-
ity, economic development, firm entry.

1 Introduction
Creativity is believed to be a key element in stimulating economic growth in cit-
ies, regions and advanced capitalist economies in general. Creativity might also
be particularly useful in knowledge-based economies, where creativity is required
to convert scientific and technological knowledge into market value. Research
into the significance of creativity increased greatly with the work of Richard Flor-
ida (2002a). His creative class focuses on the presence of people in creative pro-
fessions. A large creative class leads to a social climate with a high acceptance of
minority points of view, and improves the attractiveness of a region as a place
for highly educated people to live. These factors make an area more attractive
for innovative firms and facilitate the innovativeness of organizations. As a con-
sequence a better economic growth is anticipated. Florida and colleagues (Florida
2002b; Lee, Florida & Acs, 2004) published several articles to demonstrate these

Mainly due to Florida´s work, policy makers nowadays have much attention for
the presence of creative people and firms in their regions. The extent to which
cities and regions accommodate creative people and firms is believed to be a
major antecedent of future economic success. Especially at the level of munici-
palities and provinces, policy makers have embraced this school of thought to
develop polices to attract and develop creative firms and people in creative pro-
fessions. In the Netherlands, examples include the cities of Amsterdam, Rotter-
dam, Leiden and Eindhoven (Rutten, Manshanden, Muskens & Koops, 2004;
Metz, 2005). At the country level, policy makers have more recently begun to
develop policies oriented towards creativity. One example is the Dutch Innova-
tion Platform, a thinkthank of politicians, policy makers, entrepreneurs and busi-
ness representatives, which has declared creative industries as one of its key ar-
eas to stimulate innovation (Bekkers & Van Tilburg, 2004). Another example is
so-called Culture and Economy Bill developed by the Dutch ministries of educa-
tion and economic affairs. It proposes new policy initiatives oriented towards
creative industries (Raes & Hofstede, 2005).

One caveat is that policy makers often assume that creative industries are bene-
ficial while solid empirical evidence is still scarce. Studies that have replicated
Florida´s empirical findings in other contexts, i.e. countries outside the United
States, are still few (Boschma & Fritsch, 2007). Another caveat is that creative
professions and industries are often treated homogeneous, as if for example the
presence of artists, architects or rather game developers in IT service industries
have similar impacts on economic performance. As a consequence, policy makers
refer to Florida to legitimate a plethora of policy interventions oriented to various
industries, including visual arts, IT services, design, media firms, etc (Metz,

The purpose of this paper is twofold. First, we compare creative and non-creative
industries in a context outside the United States. Previous work on the impact of
creative industries has been done mainly in US cities, but more research in other
contexts is recommended (Levine, 2004; Braaksma, De Jong & Stam, 2005).
More specifically, we will analyse the relationship between regional shares of
creative industries and firm entry rates. Second, we analyse if firms in creative

industries are heterogeneous in terms of their business processes and contribu-
tions to the economic development of regions. In section 2, drawing on the cur-
rent literature we distinguish four creative domains: arts, media and entertain-
ment (M&E), creative business services (CBS) and knowledge intensive business
services (KIBS). We will elaborate on why we expect firms in these domains to
be heterogeneous in terms of business processes and impact on economic out-
comes. Next, section 3 analyses survey data of 4,746 Dutch SMEs to empirically
investigate the heterogeneity of business processes of firms in the four domains.
Section 4 continues with an analysis of the impact of creative industries on firm
level entry rates. This is done at the level of 457 Dutch municipalities using data
from the Dutch trade register. The paper ends with section 5 with a discussion of
our findings and suggestions for further research.

2 The nature of creative industries
Def i ni ti on and cl assi f i cati on
One of the first definitions of creative industries was established in the UK by the
Creative Industries Task Force of the Department for Culture, Media and Sport
(DCMS, 1998). Creative industries were defined as `those activities which have
their origin in individual creativity, skill and talent and which have the potential
for wealth and job creation through the generation and exploitation of intellec-
tual property´. Caves (2000, p. 1) defines creative industries as `industries that
supply goods and services that we broadly associate with cultural, artistic, or
simply entertainment value. They include book and magazine publishing, the vis-
ual arts (painting and sculpture), the performing arts (theatre, opera, concerts,
dance), sound recordings, cinema and TV films, even fashion and toys and
games´. A more recent study in the Netherlands (Rutten et al., 2004) defines
creative industries as those sectors in which goods and services are produced
that are the result of creative labor. Content and symbolic value are important
elements in this definition. In that respect, creative industries play an important
role in the development and maintenance of lifestyles and cultural identities in
society. The creative industries essentially supply informational media content
that requires `artistic creativity´ as a quintessential knowledge-based and labour-
intensive input (Handke, 2007). The cultural products they supply serve aes-
thetic, broadly educational or entertainment purposes rather than any immediate
`technical´ function (Throsby, 2001, p. 4).

Although Florida´s (2002a) creative class focuses on creative individuals, in the
current study we use data at the level of firms in creative industries. A problem
for any empirical analyses is that data on the level of individuals are difficult to
obtain. As an alternative, many empirical studies focus on creative industries
rather than people in specific professions (Caves, 2000; Turok, 2003; Gibson &
Kong, 2005; Jayne, 2005). It is easier to analyse these creative industries as
such data can be retrieved from secondary databases (that usually focus on in-
dustries rather than professions).

In the past years many theorists asked themselves what group of industries best
reflects Florida´s (2002a) creative class. The arts domain and media and enter-
tainment (M&E) industries should, without any doubt, belong to the group as the
nature of the work in these industries clearly reflects Florida´s (2002a) core of
super creative professions. This group consists of professionals for which creativ-
ity is a must, for example in architecture and design, education, arts, music and
entertainment. A classification by Rutten et al. (2004) also includes creative
business services (consisting of architects, commercial and interior designers,
etc), in addition to the arts and the media and publishing industries. As this clas-
sification also captures a large part of Florida´s creative professionals, we
adopted it for the present study. Using similar reasoning a plea can be made to
also include knowledge-intensive business services like consultancy and software
firms. This can however be criticised because the nature of work in these indus-
tries is not necessarily creative. The creative content of consultancy and IT ser-
vices is open to discussion because they do not necessarily produce goods or
services with content and symbols that are meaningful for users (Manshanden,
Raspe & Rutten, 2004; Rutten et al., 2004). Basically, creative industries can be

schematised as a series of expanding concentric circles as represented in figure
1 (Jacobs, 2005).
figure 1 Classification of creative industries
Arls Arls Arls Arls
Visual arls
Perforning arls
íedia & íedia & íedia & íedia & Enlerlainnenl Enlerlainnenl Enlerlainnenl Enlerlainnenl
Filn, radio,
TV, press
Crealive Crealive Crealive Crealive business services business services business services business services
Adverlising Technical
Knowledge inlensive services Knowledge inlensive services Knowledge inlensive services Knowledge inlensive services

The proposed classification entails four creative domains. At the core we find the
arts domain, consisting of both visual and performing arts (artists, theatre com-
panies, etc). There is no doubt that these are at the heart of the creative indus-
tries. Also at the core we find media and entertainment firms (e.g. photogra-
phers, broadcasters, journalists, publishing houses). These two domains also re-
flect Caves´ (2000) definition. Next, creative business services (CBS) such as
technical designers and advertising firms are considered to be another creative
domain. These firms employ many creative professionals with jobs of a clearly
creative nature (e.g. designing a house, creating an advertising campaign). Be-
sides, other recent studies have also regarded advertising and design as a part
of the creative industries (e.g. Grabher, 2002). In the periphery of figure 1 we
find knowledge intensive business services (KIBS) for which, as indicated above,
the creative content can be discussed.

Any demarcation of creative industries is, of course, somewhat arbitrary. This is
immediately clear from previous estimates of the share of creative industries
within the population of private firms. In the Netherlands, Braaksma et al.
(2005) found that, depending on one´s definition, creative industries comprise
between 1.7% (just the arts domain) and 19% of the business population (in-
cluding knowledge intensive business services). Given previous classifications
and debates we anticipate that figure 1 gives a broad definition of creative in-
dustries. To account for potential criticism that KIBS might be misclassified, our
empirical excercises presented hereafter are done in such a way that the conse-
quences of including KIBS can be judged. Annex II in this paper lists relevant
NACE-codes for the four domains used for our empirical exercises to be dis-
cussed later.


Potenti al heterogenei ty
Previous work suggests that creative firms probably can not be treated as a ho-
mogenous group. Rutten et al. (2004) discuss differences between firms in the
arts, media and entertainment and creative business services with regard to
their economies of scale, sensitivity to business cycle fluctuations, sources of fi-
nancial capital (credit and/or subsidies), opportunities in innovation, and mana-
gerial constraints. Likewise, the Dutch consultancy organization Syntens (2005),
a specialist in giving advice to entrepreneurs on innovation topics, has identified
similar differences, also for KIBS. Drawing on this work table 1 summarizes dis-
tinctive features of the four domains. A disclaimer is that this characterisation is
only tentative; it is based on observations in the Netherlands and might not ap-
ply to other countries (see Izushi & Aoyama, 2006; Lorenzen, 2007).
table 1 Features of creative firms in four domains
Feature Arts M&E CBS KIBS
Market focus
Joint production
and authorship
Customer focus
Functionality of
Applied creativity
Customer focus
Repertory of ser-
Combination of
Share of sub-
sidies in total
Very high Partially Small Small
Main custom-
Governments Private consum-
Large businesses Large businesses
Both individual
and collective
User does not
influence output
Complex, large-
Users of IT
Open culture,
much coopera-
Influence of busi-
ness cycles
Flexible assign-
ments of employ-
Application of
Influence of busi-
ness cycles
Output Unique products
or small series
Standardised Tailor made Tailor made within
specific knowledge
Source of in-
novation for
Hardly ever Occasionally Frequently Very often

In the arts domain, artistic performance is valued much more than commercial
performance. This domain is very dependent on subsidies. Dutch arts firms are
usually very much oriented towards their own supplies: a piece of work should
satisfy the creator´s artistic standards, and customers should not infere with this
process, but just consume. The Dutch media domain is very dynamic, with many
new entrants and collaborations and, with the exception of the public broadcast-
ing organizations and parts of the movies industry, the market mechanism domi-
nates here. Creative business services work on a commercial base; they offer
tailor-made solutions to their customers. Firms in architecture or technical or
commercial design typically operate in business-to-business environments in
which customers should be satisfied. Their activities are very innovative, either
directly or by enabling innovation by their clients. Knowledge intensive business
are a bit similar, but one major distinction is that the application of knowledge is
an essential feature of their business models.

The features in table 1 imply that firms in four domains differ in terms of their
dominant ideology, dependence on subsidies, types of customers, modes of pro-
duction, nature of outputs, and contributions to other firms´ innovation proc-

esses. This suggests that creative firms are dissimilar in their organization of
business processes, and possibly also in their contribution to economic develop-
ment. Florida himself for example regards creativity to be connected with inno-
vation opportunities which in turn contribute to growth (Florida, 2002a). For the
Netherlands, there are also empirical figures which suggest differences between
the four domains (see table 2).
table 2 Development and structure of creative firms in four domains
Feature Arts M&E CBS KIBS
All in-
Average change (L no. of
firms in two-year periods):
- 1994-1996 6.8% 6.8% 10.2% 12.5% 9.6% 4.2%
- 1996-1998 5.5% 3.6% 9.1% 13.2% 8.8% 5.9%
- 1998-2000 5.0% 2.4% 7.1% 12.9% 8.0% 0.3%
- 2000-2002 4.7% 1.3% -2.4% 5.0% 1.9% -2.7%
- 2002-2004 3.1% -3.9% -3.3% -2.7% -1.8% -1.2%

Number of firms in 2004: 18,100 8,300 30,500 39,500 96,400 642,900
- no employees 63.6% 65.9% 68.5% 73.8% 69.5% 53.6%
- 1-9 employees 33.3% 27.6% 25.6% 22.1% 25.8% 37.6%
- 10-99 employees 2.9% 5.5% 5.6% 3.8% 4.3% 8.1%
- >=100 employees 0.3% 0.9% 0.3% 0.3% 0.4% 0.7%

Employment in 2004 (* 1,000
persons): 67.0 63.9 161.6 219.1 511.6 5,541.2
- >= 12 hour per week 49.3 57.4 147.5 200.7 454.9
- < 12 hour per week 17.7 6.5 14.1 18.4 56.7
Source: Braaksma et al. (2005).

In a descriptive study in the Netherlands thas was done by Braaksma and col-
leagues (2005) it became evident that the number of firms in creative industries
has grown enormously in the last decade. However, the creative industries were
not immune to a recession at the beginning of the 21
century that had a severe
impact on the Dutch economy. Creative industries faced a net loss in terms of
numbers of firms in the 2002-2004 period. One exception was the arts domain,
which contains basically organisations depending on public funding and which
can, as a consequence, operate pretty much independent of business cycle dy-
namics. In contradiction, during the favorable conditions of between 1994 and
2000 creative business services and KIBS were able to grow at faster pace than
firms in the arts and media and entertainment domains.

The table also shows that in 2004 creative industries in the Netherlands con-
sisted of 56,900 firms; about 9% of the Dutch business population. The majority
of these businesses are self-employed entrepreneurs. Creative industries are
dominated by relatively small-scale activities, most likely because (firm-internal)
economies of scale are hard to achieve in their creative activities, and because
their work is usually labour-intensive (Canoy, Nahuis & Waagmeester, 2005).
Only 5% of the businesses in creative industries employ ten or more employees,
compared to almost 9% of the Dutch business population in general (Braaksma
et al., 2005).

The figures in table 2 also confirm some of the differences between the four do-
mains mentioned above. One example is the relatively large share of part-time
workers in the arts domain (17.7/67.0 = 26%). It suggests that working in the
arts domain often does not constitute the primary source of household income.
The arts domain has a reputation of consisting of many entrepreneurs that are
more oriented towards artistic than to business values (cf. Rutten et al. 2004).

In the other domains shares of part-timers are much lower. In the next sections
we empirically explore our presuppositions that firms in creative industries differ
in their use of business processes and contribution to economic development.

3 Heterogeneity of business processes
To explore the heterogeneity of business processes of firms within the four do-
mains we analysed a database provided by EIM Business and Policy Research, a
Dutch research institute with a focus on entrepreneurship and small firm dynam-
ics. The database was compiled using a survey that measured how small and
medium-sized entreprises (SMEs) organize their business processes. This Busi-
ness Processes Survey (BPS) covered various topics, including innovation, strat-
egy and marketing, and human resources management. The survey was done in
June 2007. It was commisioned by the Dutch ministry of economic affairs be-
cause no other publicly funded data source recorded how SMEs organize their
business processes.

The BPS draws on telephone interviews rather than internet or pen-and-paper
surveys. This is to obtain better response rates. In many industries respondents
are hard to contact due to irregular working hours and general disinterest to par-
ticipate in surveys. Respondents were those who were in charge of daily opera-
tions, usually the owner/director and otherwise the general manager. For any
survey EIM tried to contact a respondent at least five times before he/she was
marked as a non-respondent.

To analyse the business processes of firms we disposed of responses of 4,746
SMEs, defined as firms with at least one but no more than 100 employees. To ar-
rive at this response initially 9,000 firms were contacted, a response rate of
53%. A total of 574 respondents were located in creative industries, while other
respondents belonged to other industries including agriculture, manufacturing,
construction, trade, hotels and restaurants, transport, financial services and
other (non-creative) services. In table 3 it is shown how our data are distributed
across type of industry and size class. Altogether we disposed of responses from
88 firms in the arts domain, 107 firms from media and entertainment, 152 from
creative business services, and 227 from knowledge intensive business services.
A comparison of the distribution of respondents and non-respondents by type of
industry indicated that there was no non-response bias present. A χ
-test be-
tween the distributions revealed no significant differences at the 5% level (p =
0.11). For size classes a similar result was found: p(χ
) = 0.61.
table 3 Number of respondents across type of industry and size class
Size class
Industry NACE-codes 1-9 em-
Agriculture 1-5 217 115
Manufacturing 15-37 (excl. 22.1) 575 465
Construction 45 204 133
Retail, cars and wholesale trade 50-52 653 420
Hotels and restaurants 55 109 57
Transport 60-64 221 187
Financial services 65-67 85 76
Creative industries
- Arts See Annex I 50 38
- Media and entertainment " 54 53
- Creative business services (CBS) " 99 53
- Knowledge intensive business services (KIBS) " 128 99
331 243
Other services 70-74, 90-93 (excl. 348 307

Size class
Industry NACE-codes 1-9 em-
creative industries)

As the commisioner of the survey had wanted to make comparisons between
specific types of industries, the sample was disproportionally stratified. In the
creative industries, compared to the general Dutch population of SMEs, arts
firms are somewhat underrepresented at the expense of media and entertain-
ment firms. Arts firms consistute 2.2% of the Dutch SME population, but in the
sample they are only 1.9% (=88/4,746). Outside the creative domains similar
differences are found, i.e. manufacturing and transport firms are overrepre-
sented while trade firms and hotels and restaurants have a relatively low cover-
age. We did not expect this to compromise our findings, yet, we computed a
weighing variable that accounted for industry and size class differences. This
variable was computed in such a way that representative estimates were ob-
tained for the population figures of the descriptives presented hereafter
. For all
analyses presented here we checked if our weighing variable had caused misin-
terpretations, i.e. all analyses were redone with unweighed data. As this pro-
vided very similar results, we here present the weighed results. Results of the
unweighed analyses are available from the authors on request.

Indi cators
Although the BPS was not specifically done for the current paper, some of its in-
dicators enable an empirical test of similarities and differences between the crea-
tive domains. The survey covered various topics, including the innovative output
and innovative inputs of firms, strategy and marketing, and some indicators for
their use of human resources practices. See table 4.
table 4 Variables used in the analysis of business processes
Indicator Description
Innovative output
 Products new to the firm Firm developed and introduced at least one product (good or service)
new to the firm in the past three years (yes=1, no=0)
 Products new to the in-
Firm developed and introduced at least one product (good or service)
new to the industry in the past three years (yes=1, no=0)
 Process innovations Firm implemented at least one new or significantly improved process in
the past three years (yes=1, no=0)
Innovative inputs
 External networks Firm uses external networks to exchange knowledge (yes=1, no=0)
 External co-operation Firm co-operates with other parties to develop innovation (yes=1,
 Dedicated innovation
Firm has employees which are occupied with innovation as part of their
daily work (yes=1, no=0)
 Development/research Firm does any development- or research activities (yes=1, no=0)
Strategy and marketing
 Documented strategy Firm has a documented strategy or business plan (yes=1, no=0)
 Emphasis on innovation Firm regards innovation as a key strategic objective (yes=1, no=0)
 Emphasis on marketing Firm regards marketing as a key strategic objective (yes=1, no=0)
 Specialised marketing
Firm has employees which are occupied with marketing in their daily
work (yes=1, no=0)
Human resources manage-

 Profit sharing/bonuses Firm uses profit sharing or bonuses as part of its reward system
(yes=1, no=0)
 Shareholding/options Firm uses options or shares as part of its reward system (yes=1,

For this purpose we obtained recent population figures from the Dutch Chamber of Commerce
trade register database (a database that registers basically all Dutch private firms).

 Periodical reviews Firm organizes periodical reviews to assess and record employees' per-
formance (yes=1, no=0)
 Training and education Firm funds a program for employee training/education to raise their
skills (yes=1, no=0)

All questions were relatively simple dichotomous indicators. We disposed of three
innovation output indicators, all very similar to the ones in the OECD´s (2005)
Oslo Manual for the construction of innovation surveys. Innovative outputs were
recorded by asking for (1) any new or significantly improved products, (2) prod-
ucts that were also new to the industry rather than just new to the firm, and (3)
new or improved business processes in the past three years. One major differ-
ence, however, is that innovation was defined more broadly. The Community In-
novation Survey defines innovation from a technological point of view: `techno-
logical product and process innovations comprise implemented technologically
new products and processes and significant technological improvements in prod-
ucts and processes´ (OECD 2005, p. 31). The BPS employs a much broader defi-
nition; innovation is defined as all implemented products and processes which
are new to the firm and the aim of which is to provide some kind of benefit (cf.
King & Anderson 2002). This definition enables the BPS to cover a wider range of
innovations in SMEs, and not just technological innovation.

For innovative inputs four indicators were available, including firms´ use of ex-
ternal networks to exchange knowledge, partnerships to develop innovations,
employment of specialised innovation workers, and involvement in any develop-
ment- or research activities. The latter question was not limited to R&D, but
could also include market research and testing activities. For strategy and mar-
keting one indicator measured if firms had a documented strategy. Next, respon-
dents were asked if they regarded innovation and marketing as key elements in
their strategies (no matter if a strategy was documented or not). Another indica-
tor recorded if firms were employing specialised marketing workers at the time
of the survey. Finally, it was asked if firms used various human resources prac-
tices including profit sharing, option or share packages, periodical reviews and
training and education programs.

Anal ysi s and resul ts
We set up a range of multivariate analysis of variance (Anova) models to test our
presuppositions. Analysis of variance is used to uncover the effects of categorical
independent variables on an interval dependent variable (Turner & Thayer,
2001). Our analysis consisted of two steps. We first tested for any significant dif-
ferences between firms in creative and other industries. A range of Anova models
was estimated using the indicators in table 4 as dependents, and a dichotomous
variable identifying creative and non-creative firms as the independent variable.
Next, using similar models we investigated whether any differences could be re-
vealed between the four domains.

In each model we controlled for firm size. Previous studies have shown that lar-
ger firms tend to realise innovative output more often, are better at securing in-
novation resources (e.g. Vossen, 1998; Nooteboom, 1994), and are more likely
to formulate strategies and adopt human resources practices (e.g. Churchill &
Lewis, 1983). We anticipated the prevalence of the surveyed business practices
is better in larger firms, a suspicion that was confirmed by simple comparisons of
means between small- and medium-sized firms (no output shown here). We used

the log transformed number of employees as a covariate, the reasoning being
that one additional employee probably has a greater impact on the prevalence of
business processes in smaller firms than in larger ones.

All analyses were done with the Unianova procedure in SPSS. Results for the
comparisons of creative and other industries are presented in table 5. As one
could argue that KIBS are not creative industries, the table actually presents two
comparisons. On the left-hand side creative industries are defined broadly in-
cluding KIBS (as visualised in figure 1). On the right-hand side KIBS are classi-
fied in the reference group.
table 5 Prevalence of business processes in creative and other industries

incl. KIBS
Other in-
Creative in-
excl. KIBS
Other indus-
Innovative output
 Products new to the firm 50% 29% 100.4** 47% 31% 40.8**
 Products new to the industry 24% 14% 38.9** 19% 15% 3.8
 Process innovations 63% 53% 30.5** 64% 54% 21.4**
Innovative inputs
 External networks 60% 40% 116.1** 53% 42% 20.1**
 External co-operation 58% 29% 218.6** 55% 31% 84.8**
 Dedicated innovation workers 60% 40% 121.3** 52% 42% 18.5**
 Development- or research
46% 24% 161.4** 38% 26% 33.1**
Strategy and marketing
 Documented strategy 51% 39% 39.0** 45% 40% 5.0^
 Emphasis on innovation 64% 49% 16.1** 61% 50% 22.2**
 Emphasis on marketing 51% 51% 0.8 54% 51% 4.5^
 Specialised marketing work-
53% 38% 54.2** 48% 40% 7.9*
Human resources management
 Profit sharing/bonuses 32% 22% 14.8** 32% 23% 8.2*
 Shareholding/options 6% 2% 16.0** 6% 3% 13.5**
 Periodical reviews 64% 58% 12.9** 62% 59% 2.8
 Training and education 51% 42% 17.4** 49% 43% 6.2^
Log transformed firm size as covariate; ** p < 0.001, * p < 0.01, ^ p < 0.05.

F-values confirm that firms in the creative industries use the surveyed business
practices more often than their non-creative counterparts. With the exception of
strategic emphasis on marketing, all indicators are significantly different. As our
models were specified to control for firm size, these differences seem to reflect
true differences between both types of industries. Besides, when KIBS are con-
sidered to be not creative (right-hand side of table 5) most differences remain
significant. Firms in creative industries innovate more often, pay more attention
to strategic and marketing issues, and seem more advanced in their use of hu-
man resources practices. We also note that in the right-hand models F-values
are somewhat smaller with lower significance levels. This implicitly suggests that
KIBS are a deviant group of firms, a suspicion that is confirmed by our analyses
presented hereafter.

We went on to investigate if any differences could be found between the four
domains of the creative industries. See table 6. Again, two versions of the sig-
nificance tests are presented. For the presented F-values, the left-hand column
gives the significance tests for comparisons between the four domains. The
right-hand column gives the results when only three creative domains are com-
pared, i.e. KIBS are supposed to be no part of the creative industries.

table 6 Anova models of business processes between types of creative industries

Innovative output
 Products new to the firm 56% 53% 39% 53% 4.1* 6.4*
 Products new to the industry 18% 30% 15% 31% 4.4* 3.6^
 Process innovations 62% 66% 64% 62% 0.9 0.8
Innovative inputs
 External networks 48% 52% 56% 69% 4.8* 0.1
 External co-operation 62% 55% 51% 62% 1.6 2.9
 Dedicated innovation workers 49% 61% 50% 71% 6.4** 0.8
 Development- or research
38% 51% 34% 55% 5.6* 3.4^
Strategy and marketing
 Documented strategy 44% 44% 47% 57% 2.9^ 1.0
 Emphasis on innovation 56% 82% 56% 69% 5.2* 7.1*
 Emphasis on marketing 57% 65% 49% 48% 3.9* 3.4^
 Specialised marketing work-
38% 61% 50% 60% 4.2* 2.4
Human resources management
 Profit sharing/bonuses 18% 31% 40% 33% 0.8 0.4
 Shareholding/options 14% 1% 4% 7% 4.4* 5.2*
 Periodical reviews 46% 64% 71% 67% 0.4 0.5
 Training and education 40% 49% 54% 53% 0.6 0.4
Log transformed firm size as covariate; ** p < 0.001, * p < 0.01, ^ p < 0.05.

The table shows more subtle differences between the four domains. For innova-
tive output, firms in the KIBS and media and entertainment domains more often
realise product innovations. For innovative inputs KIBS firms tend to score above
the average while especially CBS firms seem to lag behind. For strategy and
marketing, we again find media and entertainment and KIBS firms in the lead.
These differences remain significant even after controlling for the size of firms.
For the identified human resources practices it appears that firms in CBS and
KIBS are using these more often. Now it is the arts domain lagging behind.

After controlling for the size of firms, we find significant differences for 10 of the
15 business process indicators. This confirms our presupposition that across the
four creative domains, firms are not homogeneous in how they organize their
business processes. We also note that when KIBS are considered to be not crea-
tive, the differences between the three remaining domains are less pronounced
and more difficult to interpret. Only six indicators remain significant (table 6).
Apparently, KIBS firms are most deviant in terms of their business processes. In
the next section we explore if similar differences can be found with regard to the
connection between creative industries and regional firm entry rates.

4 Contribution to regional firm entry
A widely recognized propostion that is eagerly adopted by policy makers is that
clustering of creative organisations generates spin offs for the whole local econ-
omy (Scott, 2000). Clusters of creative firms are proposed to generate knowl-
edge spillovers which stimulate new business development and growth in other
industries (cf. Porter, 1990). This implies that regions with high shares of crea-
tive firms are anticipated to perform better in terms of economic development.
One of the main causal relationships that Florida (2002a) proposed is that crea-
tive regions realise better rates of firm entry, i.e. more start-ups are seen, and
more existing firms tend to move into such regions. Florida´s main reasoning is
that creativity goes together with high levels of tolerance, talent and technology
start-ups, all factors which are good for new firm entry. This in turn will generate
better economic performance.

Empirical work in the Netherlands suggests that the relationship between crea-
tive industries and economic outcome variables such as regional employment
growth, is mainly indirect (Marlet & Poort, 2005). Only for Amsterdam, the big-
gest Dutch agglomeration, a direct correlation between creative industries and
employment growth can be established (p. 34). In general, results suggest that
the connection is indirect: the presence of creative industries makes a region an
attractive residence for those employees and businesses which are most needed
to boost local economic growth. In this section we build on this previous work
with an empirical exercise in which we connect regional shares of creative indus-
tries with the entry rates of new firms. Firm entry is an intermediate variable
that is supposedly connected with better economic outcomes.

For our empirical exercise we used a large dataset provided by the Dutch Cham-
bers of Commerce. This non-profit organisation manages the Dutch national
trade register (see www.kvk.nl/english/traderegister/default.asp). All private or-
ganisations in the Netherlands are obliged to subscribe to this register. We ob-
tained a database with full inventories of all private firms which had been regis-
tered in the years 2001-2006. Except for the names of organisations the data-
base contained unique firm identification numbers, four-digit postal codes to
identy firms´ location in the country, firm classification (NACE) codes and years
of registration. Using the NACE codes in Annex II we classified each firm as be-
ing a member of the arts, media and entertainment, CBS or KIBS domain. Most
firms of course were no part of the creative industries, but were classified as

All analyses were done at the level of municipalities, a level of analysis providing
sufficient detail to enable reliable estimates. To aggregate our data to this level
we merged the database for each year with a postal code-to-municipality table.
In this year there were 457 municipalities in the Netherlands
. We based the ag-
gregation on the situation in 2006. By using just a single mapping moment for

We note that higher levels of analysis were not suitable. The Netherlands has only 12 provinces
and 40 so-called Corop regions (which are rough combinations of municipalities).

the whole time period we implicitly corrected for municipal re-classifications in
previous years. This prevented biases in our measures of firm entry (see hereaf-

Vari abl es
We investigated the empirical relationship between the regional share of firms in
creative industries (relative to the total number of firms) and the regional entry
rate of new firms. An overview of our variables is provided in table 7.
table 7 Variables used in the analysis of regional firm entry
Variable Description
Average annual entry rate of firms in municipality m, in time period 2002 - 2006
Average share of firms in creative industries (including KIBS) in municipality m, in the
time period 2001 - 2006
Average share of firms in creative industries (excluding KIBS, i.e. creative industries
narrowly defined) in municipality m, in the time period 2001 - 2006
Average share of firms in arts industries in municipality m, in the time period 2001 -
Average share of firms in media and entertainment industries in municipality m, in the
time period 2001 - 2006
Average share of firms in creative business services in municipality m, in the time pe-
riod 2001 - 2006
Average share of firms in knowledge intensive business services in municipality m, in
the time period 2001 - 2006
Qm Average number of (thousands of) inhabitants in municipality m, in the time period
2001 - 2006.

Our measure Nm
indicates firm entry. It draws on the number of new firms in a
particular municipality and year, i.e. we regarded any firm as `new´ if it was reg-
istered in one municipality in a specific year but had not been there in the previ-
ous year. In so doing the measure entails both start-up firms and existing firms
which have moved from one municipality to another. It is computed as.

% 100 x 1
, 2001
, , 2001

∆ +

m t

.where N2001,m
is the total number of private firms registered in municipality m
in 2001, , and ∆Nt,m
is the number of private firms entering municipality m in
year t.

The variable Sm
represents of the average share of firms in creative industries,
using the broad definition of creative industries including knowledge intensive
business services. It is computed as.

% 100 x

m t
m t

.where Nt,m
is the number of firms in creative industries registered, and Nt,m
is the total number of private firms in municipality m and year t.

Similarly, we defined Sm
as the average share of firms in creative industries
using the narrow definition (excluding KIBS). Moreover, Sm
, Sm
, Sm
represented the average shares in 2001-2006 of respectively the arts do-
main, media and entertainment, creative business services and knowledge inten-
sive business services. For any municipality Sm
= Sm
+ Sm
+ Sm
+ Sm

and Sm
= Sm
+ Sm
+ Sm

Our analyses controlled for the influence of the size of municipalities. We antici-
pated that firms are more willing to establish themselves in urban areas rather
than rural ones. Urban areas benefits from better supplies of skilled labor, infra-
structures, etc (Jacobs, 1961; 1969). Therefore we used Qm as a control vari-
able, reflecting the average number of (thousands of) inhabitants during 2001-
2006. For each municipality m this variable was computed as.

m t m

Anal ysi s and resul ts
Based on values for 457 municipalities in the Netherlands, descriptive statistics
are shown in table 8. We notice that in the years 2002-2006 there were major
differences between municipalities in the entry rates of new firms, varying from
8.7% to 21.5%. We stress that the shares presented in table 8 cannot be com-
pared with the descriptives shown in table 2. Knowledge intensive business ser-
vices are for example the biggest domain (see table 2), but as these firms are
unevenly spread across Dutch municipalities, the average share of KIBS at the
municipal level (see table 8) is relatively low.
table 8 Descriptive statistics at the municipality level (n=457)
Pearson correlations
Variable Mean SD Min Max

13.5% 2.1% 8.7% 21.5%
7.3% 2.5% 1.8% 20.1% 0.45**
5.7% 1.9% 1.6% 15.9% 0.36** 0.94**
1.6% 0.7% 0.2% 5.7% 0.19** 0.43** 0.51**
0.7% 1.3% 0.0% 8.9% 0.25** 0.62** 0.69** 0.30**
3.4% 2.5% 0.0% 9.2% 0.29** 0.80** 0.81** 0.05 0.29**
1.6% 1.0% 0.0% 7.6% 0.46** 0.76** 0.49** 0.13* 0.27** 0.50**
Qm (1) 35.7 57.8 1.0 742.6 0.41** 0.32** 0.28** 0.15* 0.23** 0.20** 0.28**
(1) Correlations for Qm are computed with log transformed values, ** p < 0.001, * p < 0.01.

In many municipalities creative industries are only a small part of the total num-
ber of private firms. There are however a number of bigger cities with high
shares of creative firms. The share can reach maximum values of 15% up to
20%, contingent on how creative industries are demarcated.

In table 8 we notice significant correlations between the entry rate of new firms
and the share of firms in creative industries, no matter how the latter variable is
defined (strong and significant correlation between Nm
and either Sm
or Sm
But we also see positive significant correlations with the size of municipalities.
Because of the skewness to the right of the municipality size distribution, we
used log transformed values of Qm in the correlation (and subsequent) analysis.
The significant correlation of 0.52 (p < 0.001) confirms that indeed large cities
show higher entry rates than smaller ones.

The bottom row of table 8 also shows that large municipalities have higher
shares of creative industries. So within the private sector, firms in creative in-
dustries are more attracted to larger cities than their non-creative counterparts,
a finding that was previously proposed by Hall (1998) and Florida (2005). At first
sight it seems remarkable that relatively small sectors such as creative industries
are strongly related to firm entry, but we have to take into account the possibil-
ity that the municipality size effect dominates all other effects, i.e. the effect of
creative industries could be an artifact as it may be the result of `attractiveness
of larger cities´.

To empirically explore if and how creative industries and domains contribute to
firm entry, we set up a range of regression models. Each model was specified
with the firm entry rate Nm
as the dependent variable, log transformed munici-
pality size as a control variable, and subsets of the other indicators as independ-
ents. We estimated five regression models:
- The first model entered log transformed Qm as a control variable. This gives a
baseline model that all subsequent models should exceed in terms of signifi-
cance and explained variance.
- The second model added Sm
to the equation to investigate if higher shares
of creative industries go together with better firm entry rates.
- In the third model, we added the shares of the four creative domains (Sm
, Sm
and Sm
) seperately to test for significant differences between
these domains.
- The fourth model was an alternative to the second model, now entering the
share of creative industries according to the narrow definition of creative in-
dustries (Sm
- The fifth model was an alternative to the third model. It entered the three
remaining creative domains seperately (Sm
, Sm
and Sm
). Models (4)
and (5) allowed us to assess the consequences of leaving out knowledge in-
tensive business services.

We remark that the analyses were cross-sectional. There was no need to apply
panel techniques since differences in the independent variables are particularly
relevant in the cross-sectional dimension (i.e. between municipalities), and rela-
tively constant in the time dimension (shares of creative industries are relatively
constant over the years). Moreover, we eliminated variation of growth of the
number of firms due to general business cycle and other time specific effects by
applying average growth figures over five years.

All analyses were done with OLS regression. See table 9. In order to obtain com-
parable parameter estimates we standardised all variables, i.e. standardised ef-
fect parameters are shown.
table 9 OLS regression models of firm entry rates at the municipality level (n = 457)
(1) (2) (3) (4) (5)
Standardised effect parameters:
Log Qm 0.52** 0.52** 0.52** 0.52** 0.52**
0.16** 0.18**
0.03 0.06
0.01 0.13^
Model fit:
0.27 0.38 0.42 0.35 0.36

(1) (2) (3) (4) (5)
0.11 0.15 0.08 0.09
** p<0.001, ^ p<0.05.

Model (1) again shows that large municipalities have higher entry rates than
their smaller counterparts. Due to standardisation the estimated parameter of
this model is exactly the correlation coefficient between the entry rate and mu-
nicipality size (β = 0.52). It is highly significant and reflects a strong relationship
indicates 27% explained variance).

Model (2) confirms that creative industries indeed go together with better entry
rates of firms. After the municipality size effect is controlled for, an additional
significant effect parameter of the share of creative industries (broadly defined)
emerges ((β = 0.29). It creates an additional explained variance of 11%.

In model (3) we see that the effect of the creative industries can be mainly as-
signed to firms in the arts and KIBS domains. For these two domains we find
significant and positive effect parameters, while the coefficients for media and
entertainment and creative business services are insignificant. This finding sup-
ports our presupposition that firms in various domains have different impacts on
regional economic development.

Model (4) demonstrates that even when creative industries are defined in a nar-
row sense, the correlation with firm entry rates remains. The standardised effect
parameter is now 0.23 but still significant at p < 0.001, while entering the share
of creative industries indicator Sm
adds 8% to the explained variance. This im-
plies that when KIBS firms are considered to be non-creative, the relationship
with firm entry is slightly lower, but still positive.

Finally, model (5) enters the three domains of arts, media and entertainment
and creative business services seperately. For the arts domain the result is com-
parable with model (3), but for creative business services the effect parameter is
now marginally significant, implying that the relationship between CBS and firm
entry diminishes when the KIBS domain enters into the equation.

Our analysis included with a number of checks on robustness. First, the munici-
pality level could be too narrow if the prevalence of creative industries in
neighbouring areas also affect firm entry. The Netherlands is a crowded county:
cities are characterised by small geographical distances, making it well possible
to do business with firms in adjacent cities (Marlet & Van Woerkens, 2005). We
therefore repeated our analyses with shares of creative industries defined at the
two-digit postal code level. This provides a rougher indicator which basically as-
sumes that regional firm entry depends on the share of creative industries in ad-
jacent municipalities. Second, we experimented with an alternative dependent
variable that also reflecting firm exits and not just entries of new firms. Both
analyses provided nearly identical results; their outputs can be obtained from the
authors on request.

5 Discussion
This paper aimed to contribute to the creative industries literature in two re-
spects. First, we investigated the relationship between creative industries and
regional firm entry rates. After analysing data at the level of 457 Dutch munici-
palities, we found a positive and significant connection between the two. Second,
we analysed if firms in various creative industries are heterogeneous in terms of
their business processes and contribution to firm entry. Drawing on survey data
of 4,746 Dutch SMEs, it is confirmed that firms in creative industries have differ-
ent business processes, i.e. they are more innovative, pay attention to strategy
and marketing more often, and are more likely to use human resources practices
than their counterparts from other industries. More important is that we also
found empirical evidence for our presupposition that firms in various creative
domains differ in their use of specific business processes. We demarcated the
creative industries to consist of four domains: arts, media and entertainment,
creative business services, and, in the periphery of creative industries, knowl-
edge intensive business services. For the use of innovation practices, strategy,
marketing and human resources practices, significant differences between the
four domains were found. It appeared that especially knowledge intensive busi-
ness services (KIBS) were deviant, i.e. they seem even more `professional´ in
their use of the surveyed practices. When KIBS firms were discarded from the
analyses significant differences remained, but these were less pronounced. Fi-
nally, in our analysis of regional firm entry rates we also found distinctions be-
tween the four domains, i.e. municipalities with higher shares of creative firms in
the arts and KIBS domains had better firm entry rates, while media and enter-
tainment and creative business services were not (or at most marginally) signifi-

For policy makers the implications of our findings seem at first sight straightfor-
ward. As we discussed in the introduction section, policy makers have embraced
the creative industries as a driver of economic development. Numerous policies
have been developed and implemented to attract new creative firms and to sup-
port the development of such firms. Our comparisons of creative and non-
creative industries with survey data, but also our analysis of firm entries with the
trade register data, suggests that such policies make sense. Firms in creative in-
dustries are indeed more innovative. They pay more attention to strategic and
marketing issues and seem to be more progressive in their use of human re-
sources practices. Besides, at the municipal level there is a positive connection
between the share of creative industries and the average annual entry rate of
new firms. This suggests that at least some positive economic effects can be an-

It is tempting to conclude that policy interventions to attract and develop crea-
tive industries will pay off. Policy makers should however be aware of potential
differences between various creative domains. In practice policy makers use the
creativity literature, especially Florida (2002a), to legitimate a broad range of
policy interventions. These can aim for firms as diverse as visual arts, gaming,
interior design or photography. In this paper we have argued that there are very
distinctive domains within the creative industries that differ with respect to their
dominant ideology, output markets, production features, employment growth,

sensitivity to business cycles, innovativeness, etc. For a selection of business
practices and the connection with regional firm entry, we also found empirical
evidence. Policy makers should therefore not treat the creative industries as a
homogeneous group, but should rather be aware that a positive impact on eco-
nomic development might be obtained from only a subset of creative industries.

Given our findings one might tentatively conclude that firms in arts and knowl-
edge intensive business services are relatively important objects for policy de-
velopment. We are however aware that it is too early for such recommendations.
Our analysis of firm entry rates is just one in a broad range of mechanisms link-
ing creativity with economic performance. One alternative explanation would for
example be that firms in creative industries contribute to the innovativeness of
organizations in other domains, which in turn boosts regional economic perform-
ance. For creative business services (consisting of architects, technical and
commercial designers, etc) such a contribution is not hard to imagine. The inno-
vativeness connection could also be indirect, e.g. high shares of firms in creative
industries might induce knowledge spillovers (cf. Porter, 1990). Future research
should investigate much more thoroughly the similarities and differences be-
tween various creative domains, as well as their consequences for economic per-
formance. We end this paper by suggesting that such future work would be much
more valuable if researchers manage to collect primary data. Most previous work
necessarily sticked to secondary sources; the current paper was no exception to
this. Primary data would enable tests of much more sophisticated models.

Annex I References
Bekkers, F. & R. van Tilburg (2004) Voorstellen Sleutelgebieden-aanpak. Am-
bitie, excellentie en actie. Van dijkgraaf tot art director: voorstellen tot actie
van het Innovatieplatform. Innovatieplatform: Den Haag.
Boschma, R.A. and Fritsch, M. (2007): `Creative class and regional growth - em-
pirical evidence from eight European countries´. Mimeo., Urban and Regional
Research Institute Utrecht, Utrecht.
Braaksma, R. M., De Jong, J. P. J. and Stam, E. (2005): Creatieve bedrijvigheid
in Nederland [Creative businesses in the Netherlands]. EIM, Zoetermeer.
Canoy, M., Nahuis, R. And Waagmeester, D. (2005): De creativiteit van de
markt: verkenning van de rol van de overheid bij creatieve industrieën [The
creativity of the market: exploration of the role of the government in creative
industries]. CPB Document 90, Centraal Planbureau, The Hague.
Caves, R.E. (2000): Creative Industries: Contracts between Art and Commerce.
Harvard University Press, Cambridge, MA.
Churchill, N. & V. Lewis (1983), The five stages of small business growth, Har-
vard Business Review, May-June, 30-50.
DCMS (1998): Creative Industries Mapping Document. Department of Culture,
Media and Sport, London.
Florida, R. (2002a): The Rise of the Creative Class:and How It's Transforming
Work, Leisure, Community, and Everyday Life. Basic Books, New York.
Florida, R. (2002b), Bohemia and economic geography. Journal of Economic Ge-
ography 2: 55-71.
Florida, R. (2005): Cities and the Creative Class. Routledge, New York.
Gibson, C. And Kong, L. (2005): `Cultural economy: a critical review´, Progress in
Human Geography 29 (5): 541-561.
Grabher, G. (2002): `The project ecology of advertising: tasks, talents and
teams´, Regional Studies 36 (3): 245-263.
Hall, P. (1998): Cities in Civilization. Pantheon, New York.
Handke, C. (2007): `Surveying innovation in the creative industries´, paper pre-
sented at the 5th International EMAEE Conference on Innovation, Manchester,
17-19 May.
Izushi, H. And Aoyama, Y. (2006): `Industry evolution and cross-sectoral skill
transfers: a comparative analysis of the video game industry in Japan, the
United States and the United Kingdom´, Environment and Planning A 38 (10):
Jacobs, D. (2005), Creativity and the economy, Background paper for the Inno-
vation Lecture 'Compete with Creativity' 2005, Ministerie van Economische
Jacobs, J. (1961): The Death and Life of Great American Cities. Random House,
New York.
Jacobs, J. (1969): The Economy of Cities. Random House, New York.
Jayne, M. (2005): `Creative industries: the regional dimension?´, Environment
and Planning C: Government and Policy 23 (4): 537-556.
King, N. And Anderson, N. (2002): Managing Innovation and Change: a Critical
Guide for Organizations. Thomson, London.
Lee, S., R. Florida & Z.J. Acs (2004), Creativity and Entrepreneurship: A regional
analysis of new firm formation. Regional Studies 38.8: 879-891.
Levine, M.V. (2004), La classe créative et la prospérité urbaine: mythes et réali-
tés. Presentation at the Centre collégial de développement de matériel didac-
tique at the INRS Institute, Université du Québec, Montréal, Canada.
Lorenzen, M. (ed.) (2007): `Special issue: Hollywood and beyond: globalization
of the film industry´, Industry and Innovation 14 (4): 349-443.
Manshanden, W. J. J., Raspe, O. And Rutten, P. (2004): `De waarde van creati-
eve industrie´ [The value of creative industries], Economisch Statistische
Berichten 89 (4434): 252-254.
Marlet, G. & Poort, J. (2005): Cultuur en creativiteit naar waarde geschat [Cultu-
re and creativity valued]. SEO/Stichting Atlas voor gemeenten, Amster-
Marlet, G. A. & van Woerkens, C. M. C. M. (2005): Tolerance, aesthetics, ameni-
ties or jobs? Dutch city attraction to the creative class. Discussion Paper Se-
ries 05-33, Utrecht School of Economics/Universiteit Utrecht, Utrecht.

Metz, T. (2005), 'Creativiteit is geld waard', NRC Handelsblad, 18-02-05.
Nooteboom, B. (1994), `Innovation and diffusion in small firms: theory and evi-
dence´, Small Business Economics, 6 (5), 327-347.
OECD (2005): Oslo Manual: Guidelines for Collecting and Interpreting Innovation
Data. Organisation for Economic Co-operation and Development, Paris.
Porter, M. (1990): The Competitive Advantage of Nations. Free Press, New York.
Raes, S & Hofstede, B. (2006): Creatief vermogen. De economische potentie van
cultuur en creativiteit [Creative power. The economic potential of culture and
creativity]. Elsevier, The Hague.
Rutten, P., Manshanden, W., Muskens, J. and Koops, O. (2004): De creatieve in-
dustrie in Amsterdam en de regio [The creative industries in Amsterdam and
the region]. TNO Strategie, Technologie en Beleid, Delft.
Scott, A. J. (2000): The Cultural Economy of Cities. Sage, London.
Syntens (2005), AntenneWijzer: Waarnemingen van Syntens adviseurs, Den
Throsby, D. (2001): Economics and Culture. Cambridge University Press, Cam-
Turner, J.R. & J. Thayer (2001), Introduction to analysis of variance, Thousand
Oaks, CA: Sage.
Turok, I. (2003): `Cities, clusters and creative industries: the case of film and
television in Scotland´, European Planning Studies 11 (5): 549-565.
Vossen, R. W. (1998), Relative strengths and weaknesses of small firms in inno-
vation, International Small Business Journal, 16(3): 88-94.

Annex II Classification of creative industries
Domain Industries
92.31 Literacy and arts creation and interpretation
74.81 Photographic activities Visual arts
92.5 Library, archives, museums and other cultural activities
92.32 Operation of arts facilities
arts 92.34 Other entertainment activities n.e.c.
92.11 Motion picture and video production
92.12 Motion picture and video distribution
92.13 Motion picture projection
92.20 Radio and television activities
92.40 News agency activities
22.11 Book publishing
22.12 Newspaper publishing
22.13 Journal and periodical publishing
22.14 Publishing sound recordings
Media & En-
22.15 Other types of publishing
Architectural and engineering activities and related tech-
nical consultancy
74.40 Advertising
and Non-
74.87 Interior and fashion design
72.21 Publishing of software IT software
services 72.22 Other software consultancy
Research and development on natural sciences and engi-
R&D ser-
vices 73.2
Research and development on social sciences and hu-
74.13 Market research
services 74.14 Business and management consultancy

The results of EIM's Research Programme on SMEs and Entrepreneurship are
published in the following series: Research Reports and Publieksrapportages. The
most recent publications of both series may be downloaded at: www.eim.net.
Recent Research Reports and Scales Papers
H200713 19-11-2007 New Ventures´ Export Orientation: Outcome And Source Of
Knowledge Spillovers
H200712 29-10-2007 SME Choice of Direct and Indirect Export Modes:
Resource Dependency and Institutional Theory Perspectives
H200711 24-10-2007 Family Orientation, Strategic Orientation and Innovation Per-
formance in SMEs: A Test of Lagged Effects
H200710 15-10-2007 Drivers of entrepreneurial aspirations at the country level: the
role of start-up motivations and social security
H200709 12-10-2007 Does Self-Employment Reduce Unemployment?
H200708 10-9-2007 Social security arrangements and early-stage entrepreneurial
H200707 11-5-2007 Competition and innovative intentions: A study of Dutch SMEs
H200706 eind maart High-Growth Support Initiatives
H200705 14-2-2007 The relationship between economic development and business
ownership revisited
H200704 2-2-2007 The relationship between knowledge management, innovation
and firm performance: evidence from Dutch SMEs
H200703 26-1-2007 Family orientation, strategy and organizational learning as pre-
dictors of knowledge management in Dutch SMEs
H200702 3-1-2007 Ambitious Nascent Entrepreneurs and National Innovativeness
H200701 3-1-2007 Entrepreneurial diversity and economic growth
H200627 21-12-2006 Motivation Based Policies for an Entrepreneurial EU Economy
H200626 19-12-2006 Export Orientation among New Ventures and Economic Growth
H200625 18-12-2006 Institutionele voorwaarden voor zelfstandig ondernemerschap
H200624 13-12-2006 Creative Destruction and Regional Competitiveness
H200623 6-12-2006 Entrepreneurship, Dynamic Capabilities and New Firm Growth
H200622 1-12-2006 Determinants of self-employment preference and realization of
women and men in Europe and the United States
H200621 1-12-2006 Is human resource management profitable for small firms?
H200620 23-11-2006 The entrepreneurial ladder and its determinants
H200619 20-11-2006 Knowledge Spillovers and Entrepreneurs´ Export Orientation
H200618 20-11-2006 The effects of new firm formation on regional development over
time: The case of Great Britain
H200617 11-10-2006 On the relationship between firm age and productivity growth
H200616 11-10-2006 Entrepreneurship and its determinants in a cross-country setting
H200615 2-10-2006 The Geography of New Firm Formation: Evidence from Inde-
pendent Start-ups and New Subsidiaries in the Netherlands
H200614 25-9-2006 PRISMA-K: een bedrijfstakkenmodel voor de korte termijn
H200613 25-9-2006 PRISMA-M: een bedrijfstakkenmodel voor de middellange ter-
H200612 25-9-2006 PRISMA-MKB: modelmatige desaggregatie van bedrijfstakprog-
nose naar grootteklasse
H200611 25-9-2006 PRISMA-R: modelmatige desaggregatie van bedrijfstakprogno-
ses naar provincie
H200610 25-9-2006 Explaining engagement levels of opportunity and necessity en-

H200609 25-9-2006 The effect of business regulations on nascent and Young busi-
ness entrepreneurship
H200608 24-8-2006 High growth entrepreneurs, public policies and economic growth
H200607 18-8-2006 The decision to innovate
H200606 6-7-2006 Innovation and international involvement of Dutch SMEs
H200605 27-6-2006 Uncertainty avoidance and the rate of business ownership across
21 OECD countries, 1976-2004
H200604 22-6-2006 The Impact of New Firm Formation on Regional Development in
the Netherlands
H200603 21-6-2006 An Ambition to Grow
H200602 21-6-2006 Exploring the informal capital market in the Netherlands: char-
acteristics, mismatches and causes
H200601 22-5-2006 SMEs as job engine of the Dutch private economy
N200520 7-3-2006 High Performance Work Systems, Performance and Innovative-
ness in Small Firms
N200519 1-2-2006 Entrepreneurial Culture as Determinant of Nascent Entrepre-
N200518 26-1-2006 Social security arrangements and early-stage entrepreneurial
activity; an empirical analysis
N200517 23-1-2006 Determinants of Growth of Start-ups in the Netherlands
N200516 23-1-2006 Entrepreneurship in the old en new Europe
N200515 23-1-2006 Entrepreneurial engagement levels in the European Union
N200514 23-1-2006 Latent and actual entrepreneurship in Europe and the US: some
recent developments
N200513 20-1-2006 Determinants of self-employment preference and realisation of
women and men in Europe and the United States
N200512 20-1-2006 PRISMA-K: een bedrijfstakkenmodel voor de korte termijn
N200511 19-1-2006 Strategic Decision-Making in Small Firms: Towards a Taxonomy
of Entrepreneurial Decision-Makers
N200510 11-1-2006 Explaining female and male entrepreneurship at the country
N200509 11-1-2006 The link between family orientation, strategy and innovation in
Dutch SMEs: a longitudinal study
N200508 11-1-2006 From nascent to actual entrepreneurship: the effect of entry
N200507 11-1-2006 Do entry barriers, perceived by SMEs, affect real entry? Some
evidence from the Netherlands